QuickBird卫星图像信息识别
|
摘要
信息识别是目前高分辨率遥感应用中的最大障碍。以株洲市Qu ickB ird图像为研究对象,将研究区分为道路、水、林地、农用地、裸露地和居民点6种地类,分别进行目视判读、计算机监督分类和非监督分类,其精度分别为98.2%、72.64%和60.71%。同时,还对研究区内的Qu ickB ird、ETM+和TM图像进行计算机监督和非监督分类对比,结果表明无论是监督分类还是非监督分类,Qu ickB ird图像的分类精度均低于ETM+和TM图像,这说明空间分辨率的提高对传统的计算机分类结果没有改善,传统的基于像元的分类技术在应用于Qu ickB ird图像时表现出严重的缺陷。因此,本文回避了像元灰度统计法,采用先将图像分割,将以像元为基础的Qu ickB ird图像转化为以对象为基础的图像,这样将研究区共分割出10 000多个对象,建立对象的面积、周长、长度、宽度、长/宽、矩形度和圆形度计算模型;根据研究区各地类特征确定特征因子阈值,模拟目视判读过程,重新对研究区进行分类,结果6种地类的综合分类精度达到91.6%,这说明基于对象的多特征分类对于Qu ickB ird图像识别有明显的改善作用。
The information identification is the most difficult in the application of high-resolution remote sensing images.This paper focused on the method of QuickBird image's information extraction.Six kinds of land cover types: road,water,forest,agriculture,nuke and residence in the study area,Zhuzhou,Hunan,were identified by visual interpretation,supervised classification and non-supervised classification respectively,and the accuracy is 98.2%,72.64% and 60.71% correspondingly.At the same time,pre-processed images of QuickBird,ETM+ and TM were identified by supervised classification and non-supervised classification respectively and the accuracy of QuickBird image classification is lower than that of ETM+ and TM.This showed that resolution's improvement couldn't raise the accuracy of classification by the method of traditional classification.This paper avoided the traditional classification based on pixel-to-pixel and applied new method of classification based on object's gray character and shape characters.QuickBird image based on pixels was changed into new image based on objects first by segmentation,then models of measuring objects' area,perimeter,length,width,length/width,rectangle and roundness were built.Six kinds of land types were classified again by computer stimulating visual interpretation in the study area,the composite accuracy of classification is up to 91.6%,this showed that the method based on objects is a very effective way to improve the accuracy of classification of QuickBird image.
The information identification is the most difficult in the application of high-resolution remote sensing images.This paper focused on the method of QuickBird image's information extraction.Six kinds of land cover types: road,water,forest,agriculture,nuke and residence in the study area,Zhuzhou,Hunan,were identified by visual interpretation,supervised classification and non-supervised classification respectively,and the accuracy is 98.2%,72.64% and 60.71% correspondingly.At the same time,pre-processed images of QuickBird,ETM+ and TM were identified by supervised classification and non-supervised classification respectively and the accuracy of QuickBird image classification is lower than that of ETM+ and TM.This showed that resolution's improvement couldn't raise the accuracy of classification by the method of traditional classification.This paper avoided the traditional classification based on pixel-to-pixel and applied new method of classification based on object's gray character and shape characters.QuickBird image based on pixels was changed into new image based on objects first by segmentation,then models of measuring objects' area,perimeter,length,width,length/width,rectangle and roundness were built.Six kinds of land types were classified again by computer stimulating visual interpretation in the study area,the composite accuracy of classification is up to 91.6%,this showed that the method based on objects is a very effective way to improve the accuracy of classification of QuickBird image.
引文
1 L in H,He A G,L i J P.H igh-resolution Remote Sensing and Appli-cation[M].Changsha:Press of Central South Un iversity,2004:5~10.[林辉,何安国,李际平.高分辨率遥感及其应用[M].长沙:中南大学出版社,2004:5~10.]
2 L in H,L iu T L,L i J P.Remote Sensing Text-book[M].Changsha:Press of Central South Un iversity,2002:65~90.[林辉,刘泰龙,李际平.遥感技术基础教程[M].长沙:中南大学出版社,2002:65~90.]
3 L in H,He A G,ZHAO Y P.Qu ickB ird image processing and appli-cation[J].Remote Sensing Information,2004,18(2):38~42.[林辉,何安国,赵煜鹏.Qu ickB ird图像处理及其应用[J].遥感信息,2004,18(2):38~42.]
4 Tang G A,Zhang Y S,L iu Y M.D igital Remote Sensing Data Pro-cessing[M].Beijing:Press of Science,2004:170~190.[汤国安,张友顺,刘咏梅编著.遥感数字图像处理[M].北京:科学出版社,2004:170~190.]
5 Zhang X C,Huang Z C,Zhao Z H.D igital Remote Sensing DataProcessing[M].Hangzhou:Press of Zhejiang Un iversity,1997:87~107.[章孝灿,黄智才,赵之洪编著.遥感数字图像处理[M].杭州:浙江大学出版社,1997:87~107.]
6 Sun J B,Shu N,Guan Z Q.Remote Sensing Principle,M ethod andApplication[M].Beijing:Press of Survey and D raw ing,1997:195~200.[孙家?,舒宁,关泽群编著.遥感原理、方法和应用[M].北京:测绘出版社,1997:195~200.]
7 M ei A X,PengW L,Q ing Q M.Remote Sensing Introduction[M].Beijing:Press ofH igh Education,2001:109.[梅安新,彭望?,秦其明.遥感导论[M].北京:高等教育出版社,2001:109.]
8 Ou Y G.ForestRemote Sensing[M].Beijing:Press ofCh ina Forest-ry,1989:114~132.[欧润贵.林业遥感[M].北京:中国林业出版社,1989:114~132.]
9 N ing X N,Lu S T,Yang X Q.Remote Sensing Image Processing andApplication[M].Beijing:Press of Seismology,1995:75~80.[宁小年,吕松棠,杨小勤.遥感图像处理与应用[M].北京:地震出版社,1995:75~80.]
10 Kenneth R Castleman.D igital Image Rocessing Castlman[M].Translate by Zhu Z Z,L in X Y,Sh i D J,et al,Beijing:E lectron icEngineering Press,2002:375~410.[Kenneth R Castleman.数字图像处理[M].朱志刚,林学訚,石定机等译.北京:电子工业出版社,2002:375~410.]
11 Zhang L J.Image Segm entation[M].Beijing:Press of Science,2001:2~12.[章毓晋.图像分割[M].北京:科学出版社,2001:2~12.]
12 CarlottoM J.Spectral shape classification ofLandsat thematic mapperImagery[J].Photogramm etric Engineering and Remote Sensing,1998,64(9):905~913.
13 D iksh it O,Roy D P.Emp irical investigation of image resamplingeffects upon the spectral and textural supervised classification of ah igh spatial resolution mu ltispectral image[J].Photogramm etric En-gineering and Remote Sensing,1996,62(9):1085~1092.
14 Ashton E A,Schaum A.A lgorithm s for the detection of sub-p ixel tar-gets in mu ltispectral imagery[J].Photogramm etric Engineering andRemote Sensing,1998,64(7):723~731.
15 Badhwar G D.C lassification of corn and soybeans usingmu ltitemporalthematic mapper data[J].Remote Sensing of Environm ent,1985,16(3):175~181.
16 B ruzzone L.Mu lti-source classification of complex rural areas by sta-tistical and neural-network approaches[J].Photogramm etric Engi-neering and Remote Sensing,1997,63(5):523~533.
17 Cortijo F J,N P D e La B lanca.Improving classical contextual classi-fications[J].International Journal ofRemote Sensing,1998,19(8):1591~1613.
18 C rist E P m,C icone R C.Application of the tasseled cap concept tosimu lated thematic mapper data[J].Photogramm etric Engineeringand Remote Sensing,1984,50(3):343~352.
19 Huang N,L iu X J,Zhu H M.Study on the classification of remotesensing image[J].Journal ofM easuring and Test of College of HuaBei Engineering,2001,13(2):86~92.[黄宁,刘小军,朱敏慧.遥感图像分类技术研究[J].华北工学院测试技术学报,2001,13(2):86~92.]
20 L i S,D ing S Y,Shu S M.Comparison of m ethods of classifying re-mote sensing image[J].Journal ofHenan Un iversity,2002,21(2):70~73.[李爽,丁圣彦,许叔明.遥感影像分类方法比较研究[J].河南大学学报,2002,21(2):70~73.]
21 Chen S,Q ing Q M.M ethod of classifying text structure of h igh-reso-lution image based on m in i-wave[J].Geography and Geograph ic Sci-ence,2003,20(3):6~9.[陈杉,秦其明.基于小波变换的高分辨率影像纹理结构分类方法[J].地理与地理信息科学,2003,20(3):6~9.]
22 Zhen J,Luo J C,Chen Q X.Integration of models and intelligentsystem of remote sensing image[J].Earth Information Science,2003,17(1):95~101.[郑江,骆剑承,陈秋晓.遥感影像理解智能化系统与模型集成方法[J].地球信息科学,2003,17(1):95~101.]
23 Zhang B G.Artificial nerve network’s application in classifying d igit-al remote sensing image[J].Rom te Sensing of Land Resource,1998,7(1):21~27.[张宝光.人工神经网络在遥感数字图像分类处理中的应用[J].国土资源遥感,1998,7(1):21~27.]
24 Luo J C,Zhou C H,Yang Y.Study on the integration of know ledgeand models ofANN in remote sensing image classification[J].Jour-nal of Remote Sensing,2001,5(2):122~129.[骆剑承,周成虎,杨艳.人工神经网络遥感影像分类模型及其与知识集成方法研究[J].遥感学报,2001,5(2):122~129.]
2 L in H,L iu T L,L i J P.Remote Sensing Text-book[M].Changsha:Press of Central South Un iversity,2002:65~90.[林辉,刘泰龙,李际平.遥感技术基础教程[M].长沙:中南大学出版社,2002:65~90.]
3 L in H,He A G,ZHAO Y P.Qu ickB ird image processing and appli-cation[J].Remote Sensing Information,2004,18(2):38~42.[林辉,何安国,赵煜鹏.Qu ickB ird图像处理及其应用[J].遥感信息,2004,18(2):38~42.]
4 Tang G A,Zhang Y S,L iu Y M.D igital Remote Sensing Data Pro-cessing[M].Beijing:Press of Science,2004:170~190.[汤国安,张友顺,刘咏梅编著.遥感数字图像处理[M].北京:科学出版社,2004:170~190.]
5 Zhang X C,Huang Z C,Zhao Z H.D igital Remote Sensing DataProcessing[M].Hangzhou:Press of Zhejiang Un iversity,1997:87~107.[章孝灿,黄智才,赵之洪编著.遥感数字图像处理[M].杭州:浙江大学出版社,1997:87~107.]
6 Sun J B,Shu N,Guan Z Q.Remote Sensing Principle,M ethod andApplication[M].Beijing:Press of Survey and D raw ing,1997:195~200.[孙家?,舒宁,关泽群编著.遥感原理、方法和应用[M].北京:测绘出版社,1997:195~200.]
7 M ei A X,PengW L,Q ing Q M.Remote Sensing Introduction[M].Beijing:Press ofH igh Education,2001:109.[梅安新,彭望?,秦其明.遥感导论[M].北京:高等教育出版社,2001:109.]
8 Ou Y G.ForestRemote Sensing[M].Beijing:Press ofCh ina Forest-ry,1989:114~132.[欧润贵.林业遥感[M].北京:中国林业出版社,1989:114~132.]
9 N ing X N,Lu S T,Yang X Q.Remote Sensing Image Processing andApplication[M].Beijing:Press of Seismology,1995:75~80.[宁小年,吕松棠,杨小勤.遥感图像处理与应用[M].北京:地震出版社,1995:75~80.]
10 Kenneth R Castleman.D igital Image Rocessing Castlman[M].Translate by Zhu Z Z,L in X Y,Sh i D J,et al,Beijing:E lectron icEngineering Press,2002:375~410.[Kenneth R Castleman.数字图像处理[M].朱志刚,林学訚,石定机等译.北京:电子工业出版社,2002:375~410.]
11 Zhang L J.Image Segm entation[M].Beijing:Press of Science,2001:2~12.[章毓晋.图像分割[M].北京:科学出版社,2001:2~12.]
12 CarlottoM J.Spectral shape classification ofLandsat thematic mapperImagery[J].Photogramm etric Engineering and Remote Sensing,1998,64(9):905~913.
13 D iksh it O,Roy D P.Emp irical investigation of image resamplingeffects upon the spectral and textural supervised classification of ah igh spatial resolution mu ltispectral image[J].Photogramm etric En-gineering and Remote Sensing,1996,62(9):1085~1092.
14 Ashton E A,Schaum A.A lgorithm s for the detection of sub-p ixel tar-gets in mu ltispectral imagery[J].Photogramm etric Engineering andRemote Sensing,1998,64(7):723~731.
15 Badhwar G D.C lassification of corn and soybeans usingmu ltitemporalthematic mapper data[J].Remote Sensing of Environm ent,1985,16(3):175~181.
16 B ruzzone L.Mu lti-source classification of complex rural areas by sta-tistical and neural-network approaches[J].Photogramm etric Engi-neering and Remote Sensing,1997,63(5):523~533.
17 Cortijo F J,N P D e La B lanca.Improving classical contextual classi-fications[J].International Journal ofRemote Sensing,1998,19(8):1591~1613.
18 C rist E P m,C icone R C.Application of the tasseled cap concept tosimu lated thematic mapper data[J].Photogramm etric Engineeringand Remote Sensing,1984,50(3):343~352.
19 Huang N,L iu X J,Zhu H M.Study on the classification of remotesensing image[J].Journal ofM easuring and Test of College of HuaBei Engineering,2001,13(2):86~92.[黄宁,刘小军,朱敏慧.遥感图像分类技术研究[J].华北工学院测试技术学报,2001,13(2):86~92.]
20 L i S,D ing S Y,Shu S M.Comparison of m ethods of classifying re-mote sensing image[J].Journal ofHenan Un iversity,2002,21(2):70~73.[李爽,丁圣彦,许叔明.遥感影像分类方法比较研究[J].河南大学学报,2002,21(2):70~73.]
21 Chen S,Q ing Q M.M ethod of classifying text structure of h igh-reso-lution image based on m in i-wave[J].Geography and Geograph ic Sci-ence,2003,20(3):6~9.[陈杉,秦其明.基于小波变换的高分辨率影像纹理结构分类方法[J].地理与地理信息科学,2003,20(3):6~9.]
22 Zhen J,Luo J C,Chen Q X.Integration of models and intelligentsystem of remote sensing image[J].Earth Information Science,2003,17(1):95~101.[郑江,骆剑承,陈秋晓.遥感影像理解智能化系统与模型集成方法[J].地球信息科学,2003,17(1):95~101.]
23 Zhang B G.Artificial nerve network’s application in classifying d igit-al remote sensing image[J].Rom te Sensing of Land Resource,1998,7(1):21~27.[张宝光.人工神经网络在遥感数字图像分类处理中的应用[J].国土资源遥感,1998,7(1):21~27.]
24 Luo J C,Zhou C H,Yang Y.Study on the integration of know ledgeand models ofANN in remote sensing image classification[J].Jour-nal of Remote Sensing,2001,5(2):122~129.[骆剑承,周成虎,杨艳.人工神经网络遥感影像分类模型及其与知识集成方法研究[J].遥感学报,2001,5(2):122~129.]
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心